I did not come up with a good worksheet last week. We did a lot of examples, but I didn’t do anything unified.

On the other hand, here are some interesting resources I drew on for examples of continuous random variables in Python.

• PMF, PDF, and CDF basics using histograms here.
• Some iPython notebooks from the Statistical Learning online course by Hastie and Tibshirani. Original code was in R.
• Very pretty visualizations and plotting of distributions using the Seaborn package. I used a lot from this page in putting together my own iPython notebook.

This week was spring break so nothing new — only relaxation! I will have to make public some of the review material I’ve been working on, though, so that students can get another view on the material before class this week. A few more discussions and then another midterm… the semester races onward!

Why Python when I’ve been concentrating on Ruby for the last year? Popularity and the way Python’s developed so robustly into a data science tool with pandas/numpy/scipy. In particular, I want to do some time series analysis I’ll write about next week for another project and Python is just worlds ahead in time series. However, Ruby is developing on the stats and data visualization fronts too, and if I can schedule my time very well maybe I can play with some of that this weekend.

We all remember, right, that blog came from weblog, which is a combination of the words “web” and “log”. Going back to the olden days of logs — records of what you’re doing — I’m going to muse about statistics for the moment. At the moment…

I just finished up some lecture note tweaking for my probability class. We’re going to look at exponential and Poisson distributions today, and normal and lognormal distributions. Despite loving Ruby more than Python I’m trying out some IPython notebook modeling of distributions for a nice way to visualize what’s going on. Python has a much different culture of documentation than Ruby and Rails, I must say, but that’s a different post. Back to the moment. I’d like to get students to work in class on some of these problems — not just doing problems, but doing problems, exploring connections, and working through proofs. I don’t have a computer-filled classroom so can’t do online activities. What would nice worksheets look like for this topic? We’ve been down this road before….. see earthcalculus.com!

Musing: primary goals for this class would include

• comfort with manipulating the normal distribution
• comfort with the relationship between normal and lognormal, and an understanding of why one comes from multiplicative effects and the other from summative effects
• a grasp of the memoryless property of the exponential distribution

If I come up with a decent worksheet I’ll post it.